Characterizing structure-function relationships is of fundamental importance across science, from biology to condensed matter physics. These relationships provide deep insights that can lead to innovation, whether it is connecting the structure of a protein to its activity or understanding how crystalline defects alter the function of a material. Conjugated polymers are a class of plastics that can be electronically conductive, and have many potential applications due to their unique blend of physical and electronic properties. In this thesis, we focus on the atomic structure of conjugated organic systems and the effect it has on their electronic properties.

Carrier mobility in conjugated polymer materials is limited by the structure of amorphous chains that connect domains of varying crystallinity and orientation. Furthermore, for a wide range of conjugated polymers, it is established that doping and excitation induce torsional rearrangements. Nevertheless, little is known about the long-range impact these rearrangements have on chain structure. To further optimize carrier mobility in conjugated polymer materials, an improved understanding of doped and excited amorphous chain structure is necessary. We develop a multiscale model that captures the underlying electronic structure with torsion potentials which are then used to generate chain conformations as a function of doping or excitation. We confirm that the ground-state torsion potential minima are non-planar and that the minima shift to planar configurations in the doped and excited states. As a result, chain planarity monotonically increases with the level of doping or excitation, and carrier mobility is fundamentally connected to planarity. However, for the model system polythiophene, increasing planarity does not always correspond to more linear or longer chains. We find that the chain length trends diverge between doped and excited chains, despite exhibiting similar planarity. Our results offer structural insights for design strategies to tune electronic properties of aromatic conjugated polymers.

Although our polymer model provides key information at the chain level, it is essential to understand the specific interactions that govern the torsional structure of conjugated systems to effectively design functional organic materials. Creating planar or "locking" molecular structures are of particular interest for tuning electronic properties. While the incorporation of noncovalent locks is an effective strategy for increasing planarity, the precise interactions leading to these planar structures are often unknown or mischaracterized. In this thesis, we demonstrate that aromaticity can be used to understand and interrupt the complex physical interactions which lead to planarity. We illustrate the important role aromaticity has in determining structure through torsional preferences, and find that common noncovalent locks increase aromaticity near planar torsional configurations. Ultimately, we identify hyperconjugation as the key stabilizing interaction that modifies aromaticity and results in planar structures. Our systematic study explains the success of prevalent noncovalent locks in conjugated molecules and polymers and will aid in the design of improved materials for organic electronics.

Ab initio methods provide a powerful tool in the search for novel polar materials. In particular, there have been efforts to identify lead-free piezoelectric material systems to re- place PbZr0.52Ti0.48O3. This work first utilizes the Materials Project database of piezoelectric tensors to develop a design rule for disorder-tolerant piezoelectric materials based on the presence of multiple stable optical phonon modes which contribute to the polar response. This criteria is used to produce novel materials with large computationally predicted response. A methodology for exploring alloy systems utilizing a Vegard’s law-like linear interpolation of properties is then developed to capitalize on these identified disorder-tolerant materials. A parent ferroelectric Sr2Nb2O7 compound is chosen as a parent material for which all reasonable isovalent cation substitutions are considered and explored. Based on our approximations, thermodynamic analysis and density-functional theory (DFT) validation of the large scale system, Sr2Nb2−2xV2xO7 arises as a promising polar system. Sr2Nb2−2xV2xO7 is synthesized as single-crystalline thin-film heterostructures using pulsed-laser deposition and an enhanced dielectric response is observed at x = 0.05 and x = 0.1. This alloy and methodology are presented to aid the search for additional lead-free piezoelectric systems to replace PbZr0.52Ti0.48O3 (PZT). Finally we present a software package, Automated X-Ray Diffraction to Structure (AXS), to aid in the materials discovery bottleneck of characterization. The algorithm is based solely on symmetry considerations, and an efficient structure generation algorithm coupled with DFT. AXS is benchmarked against a diverse dataset of structures for which is solves 92% of cases and is also demonstrated to provide ground state structures for showcased experimental systems.

Most of the battery electrolyte candidates for multivalent-ion systems (e.g. Mg, Ca) present interesting physicochemical properties. In these are included anomalous conductivity, where molar conductivity increases with salt concentration. This phenomenon is related to the electrochemical performance and electrolyte viability. In this thesis, this phenomenon is explained via understanding and quantification of the strong inter-ionic interactions in solution. To this end, computational methods including classical molecular dynamics and electronic structure methods are combined with dielectric relaxation spectroscopy measurements. Two cases studies with ether based electrolytes, one with MgTFSI2 salt and one with Ca(BH4)2 are undertaken. In both cases, the anomalous conductivity originates from changes in salt speciation, i.e. changes in the population of distinct chemical species formed by the ions in the solution. A larger fraction of ionic salt species are promoted at higher salt concentration at the expense of larger, associated clusters. Moreover, it is found that the increase in dielectric constant with increasing salt concentration, due to a significant fraction of polar associated salt species, is likely one of the thermodynamic driving forces for this effect. The findings here are important for the rational design of multivalent electrolytes for electrochemical applications, and help inform salt and solvent selection for advanced electrolyte formulations.